Main Page
Deanship
The Dean
Dean's Word
Curriculum Vitae
Contact the Dean
Vision and Mission
Organizational Structure
Vice- Deanship
Vice- Dean
KAU Graduate Studies
Research Services & Courses
Research Services Unit
Important Research for Society
Deanship's Services
FAQs
Research
Staff Directory
Files
Favorite Websites
Deanship Access Map
Graduate Studies Awards
Deanship's Staff
Staff Directory
Files
Researches
Contact us
عربي
English
About
Admission
Academic
Research and Innovations
University Life
E-Services
Search
Deanship of Graduate Studies
Document Details
Document Type
:
Thesis
Document Title
:
LOGICAL QUALITY ASSURANCE ASSIGNMENT FOR HETEROGENEOUS TASKS IN CROWDSOURCING SYSTEMS
التعيين المنطقي للمهام غير المتجانسة مع ضمان الجودة في أنظمة التعهيد الجماعي
Subject
:
Faculty of Computing and Information Technology
Document Language
:
Arabic
Abstract
:
Recently, crowdsourcing systems have gained a huge popularity. Quality control has become rather a necessity, although it is a challenging issue, due to the unreliability, diversity or lack of skills in the crowd. This thesis aims to find a quality assurance algorithmic crowdsourcing solution, which maximizes the overall quality of the crowdsourced output while taking into consideration tasks diversity and human factors. The proposed solution includes two main components: data modeling and task assignment. This thesis proposes a quality assurance crowdsourcing data model (QADM) that considers both diversity of worker skills and changing behavior. The goal is to appropriately model the worker quality and do so by effectively modelling worker suitability for new tasks, worker reputation, worker accuracy in completed tasks and worker expected accuracy in new tasks. To improve the worker accuracy estimation, a Task-to-Task Similarity algorithm is developed that achieves higher accuracy than Cos(topic), Cos(tf-idf) and Jaccard similarity methods. The quality assurance task assignment decision problem is solved as a binary classification problem. The results achieved show that QADM exhibit accuracy improvement by 0.6 over the baseline model and over the CrowdAdvisor model by 0.005 and 0.013 in Decision Tree and SVM classifiers, respectively. The ingredients of the QADM are used as objectives to the task assignment problem, which is formalized as a multi-objective optimization problem and it is proven to be NP-hard. A memetic-based quality assurance task assignment algorithm is proposed. It applies both Differential Evolution (DE) and Large Neighborhood Search (LNS) in a way to better make use of the exploration ability of DE as a global search and the exploitation ability of LNS as a local search. The experimental results show that the using a simple DE algorithm with the large neighborhood search as a local search during the evolution cycle yields the best performa
Supervisor
:
Dr. Ahmad A. Alzahrani
Thesis Type
:
Master Thesis
Publishing Year
:
1441 AH
2019 AD
Co-Supervisor
:
Dr. Seyed M Buhari
Added Date
:
Sunday, December 15, 2019
Researchers
Researcher Name (Arabic)
Researcher Name (English)
Researcher Type
Dr Grade
Email
أروى شاكر بخاري
Bokhari, Arwa Shaker
Researcher
Master
Files
File Name
Type
Description
45674.pdf
pdf
Back To Researches Page